Study design
A Phase II open-label randomized controlled trial comprising two groups: (i) Intervention (Dignity Therapy offered in addition to standard psychological care), and (ii) Control group (standard psychological care). Consenting participants have been randomly allocated to one of these two groups after baseline (T1) measures have been collected. Standard psychological care will vary between care homes. None are providing Dignity Therapy. The nature of such care is being collected as part of this study.
Randomization
Randomization was conducted by an independent statistician. Treatment allocation (Dignity Therapy or control) was performed by block randomization with a fixed block size of two. Allocation concealment is facilitated by using sequentially numbered opaque sealed envelopes for consecutive and eligible participants. To reduce the risk of bias, the research assistant opens the next envelope to ascertain which group the resident has been allocated to after baseline measures have been collected from participants.
Ethical Approval
This study has been approved by The Joint South London and Maudsley and the Institute of Psychiatry NHS Research Ethics Committee (24/11/2008, ref: 08/H0807/75), and meets the requirements of the local Research Governance Framework.
Participants
The sample will comprise 64 residents in care homes in South East London (Lambeth, Southwark & Lewisham). Based on our recent experiences of recruiting residents of care homes, we expect to recruit approximately 10 residents in each care home. Consequently we have opportunistically recruited six care homes to the study. These are local homes in which we have conducted research in the past which expressed an interest in Dignity Therapy. We will approach more care homes if this becomes necessary. Characteristics of care homes (size, ownership, registration, standard care) will be recorded and reported.
Inclusion criteria
Residents aged 65 years old or over are included. Not all residents have a "terminal" illness and they are not selected on the basis of receiving palliative care. However, all residents in nursing homes are frail and could be considered as reaching the end of life phase. Participants are not screened for spiritual or psychological distress, or loss of dignity, however, these are assessed at baseline to explore the potential moderating effects of these variables on the impact of the intervention.
Exclusion criteria
Residents are excluded if care home managers feel they are too ill to be interviewed, or unable to provide informed consent either due to cognitive problems, or to the severity of their illness, or because they are unable to understand English. Residents with moderate or severe cognitive impairment are excluded since their "generativity" documents are likely to reflect a fractured sense of self, which could be distressing to them and recipients of these documents.
Recipients of the Dignity Therapy documents (usually family or friends) are invited into the study to give their views on taking part in the study and the therapy.
Sample size calculations
We are obtaining in-depth information on taking part in the study and receiving therapy from a relatively small sample, rather than aiming to detect significant group differences. One of the aims of the study is to estimate the effect size for an RCT. Allowing for a 25% loss at follow-up, a sample of 64 will give us 77% power to detect an effect size of 0.80 standard deviation units in the Dignity Inventory [
9] between the intervention and control groups at the (two-tailed test) 5% level of significance. This will be sufficient to estimate recruitment, follow-up rates and the sample size for a Phase III trial, and to explore qualitative outcomes.
Intervention group
The therapy is delivered by a palliative care nurse who has been trained in Dignity Therapy by Harvey Chochinov (who developed it). Training included the theoretical basis for the intervention, demonstrations of Dignity Therapy, a detailed overview of the manual, editing the Dignity Therapy documents and working with residents to produce a document that will be helpful for them and its recipients.
A standard framework of questions used in the therapy is given to residents in the intervention group (after randomisation) to give them the opportunity to think about their responses before the session (Table ). The question framework provides a flexible guide for the nurse delivering the therapy (the therapist) to shape the interview, based on patients' level of interest and responses. The therapist follows the residents' cues, helping them to structure and organise their thoughts, for example, by asking questions about time sequences, how events are causally related to each other and facilitating the disclosure of thoughts, feelings and memories. These interviews are tape-recorded, quickly transcribed verbatim then shaped into a narrative using a formatted editing process. This includes clarifications (eliminating colloquialisms, non-starters and sections not related to the "generativity" material, such an interruptions), chronological corrections, tagging and editing any content that might inflict significant harm on recipients of the document (after discussion with the resident) and finding a suitable ending for the document which is appropriate to the residents' overall message. Another session is arranged for the therapist to read the edited transcript to the participants, who are invited to make any editorial suggestions, including identifying errors of omission or commission. Once these "generativity" documents are finalised, they are given to residents to share with anyone they choose. There is no evidence from previous studies that this has been problematic.
| Table 1Framework of questions used in Dignity Therapy |
Since residents' conditions can fluctuate rapidly, the timing of the contacts can be relaxed and meetings rescheduled. If a resident's condition deteriorates, meetings are rescheduled up to three times before sensitively withdrawing them from the study. The therapist makes detailed notes of her experiences of giving each intervention and any deviations from the protocol. One in three therapy transcripts is randomly selected for review by the principal investigators. A quality assurance protocol has been developed to assess adherence to the Dignity Therapy protocol and deviations from the protocol will be reported as part of the feasibility study.
Control group
Residents in the control group have at least three interviews with the research assistant. Completing the measures and taking part in the interview gives them an opportunity to talk about their feelings. The extent to which they feel that this is therapeutic is explored in the interviews.
Recruitment Procedure
The recruitment and follow-up procedure is shown in Figure . Care home managers are asked to distribute patient information sheets and reply slips (indicating whether or not they are interested in taking part in the study) to all residents eligible for the study. At least one week is given for full consideration and discussion of participation in the study with family and friends. Residents are asked to give their reply slips to the care home manager who gives them to the research assistant. The research assistant organises a convenient time to meet with the resident. At this meeting the research assistant: (i) explains the study to the resident, (ii) answers any questions they may have about their participation in the study, (iii) checks that they have fully understood the remit and implications of the study before obtaining written consent and proceeding with the Time 1 (baseline) interview. As a final check residents are screened with the Blessed Orientation Memory Concentration Test [
10] to assess cognitive functioning. It has been suggested by Chochinov (personal communication) that Dignity Therapy is not suitable for residents with a score equal to or greater than 15. In such cases residents are excluded. This is done sensitively: the research assistant spends some time chatting with them about neutral topics. This approach has worked successfully in a previous study involving residents of care homes [
7]. The proportion of residents excluded at this stage will be reported.
The research assistant collects baseline data from all residents taking part in the study (T1), then opens the next envelope in sequence to ascertain group allocation. Residents in the control group are reminded that they will still have the opportunity to talk to the researcher about how they are feeling and about their views on taking part in the study. Residents in the intervention groups are given the framework of therapy questions (Table ) so that they have an opportunity to think about their answers before the therapy interview. The research assistant arranges a time for the therapist to conduct the therapy interview (usually within a couple of days). She also leaves a picture of the therapist with the participant information sheet and the date and time of the therapy session (session reminder card) so that the resident is not confused when visited by another person. The therapist arranges a mutually convenient time for the therapy visits. Although these documents can be bequeathed once the participant has died, they can also be given to people of their choosing at any time. If they wish, residents can choose to keep the document for themselves and not bequeath it. Care home managers keep a list of all the residents who have "generativity" documents to ensure they are sent to recipients once the resident has died. The nurse leaves a session reminder card with the resident as a reminder of when the research assistant will call back to arrange the T2 interview (approximately one week later).
At recruitment the research assistant asks residents in the intervention group to nominate a family member or friend who visits them regularly in the care home. With the resident's permission, the nominated family member or friend is invited to take part in a telephone interview once Dignity Therapy has been completed. The nominated family member or friend is often the recipient of the "generativity" document. Residents are not excluded if they do not nominate a family member. Should they wish to do so, they can chose to leave their "generativity" document with their personal effects rather than to a named person. If the resident mentions that they wish to bequeath to someone other than the nominated family member or friend we will seek permission from the resident to also invite them for interview. The research assistant follows up with the care home to determine when the resident has died. We complete interviews three months post bereavement with all nominated family and friends and any additional recipients of the document.
Outcome measures
Both quantitative and qualitative outcomes are assessed. These are collected from residents by face-to-face interviews at two time-points: approximately one (T2) and eight weeks (T3) after the dignity document has been completed, and the equivalent in the control group. Longer-term follow-up has not been included in the previous evaluation of Dignity Therapy, or in the current trials being conducted by Chochinov and colleagues. However, we believe it is important to evaluate the longer-term impact of the intervention. The feasibility of doing this is explored as part of this feasibility study. Quantitative measures have been validated and used with older people and are fairly quick and easy to use. Qualitative interview schedules have been developed for the study.
Main outcome for residents
The primary outcome is residents' sense of dignity (potential effectiveness). This will be assessed at baseline (T1) and at one week (T2) and eight weeks (T3) follow-up, using the Patient Dignity Inventory [
9]. This measure evolved directly from the dignity model, therefore, questions correspond to each of the model themes and sub-themes, including: physical, psychosocial, existential and spiritual domains of concern or distress. This measure has been validated in Canada and has been shown to have excellent face, internal, test-retest and concurrent validity.
Secondary outcomes for residents
Potential effectiveness is also assessed using: the Geriatric Depression Scale [
11] (psychological distress); the Herth Hope Index [
12] (hopefulness); the EQ-5D [
13], and a two item measure used by Harvey Chochinov in the current trials of Dignity Therapy (quality of life). With the consent of participants, we are conducting a detailed qualitative analysis of the therapy transcripts to provide insight into concerns which might impact on the effectiveness of the intervention.
To assess feasibility of delivering Dignity Therapy in this setting, time taken to organize and conduct the Dignity Therapy sessions, transcribe and edit narratives, deviations from the therapy protocol and uncompleted interventions and the reasons, and the therapist's perceptions of competence as a result of training are recorded.
To assess the
acceptability of Dignity Therapy we are conducting semi-structured interviews with residents in the intervention group to obtain their views on the intervention. The therapist is recording her experiences of delivering the therapy and her observations of resident's responses during and after the therapy. As suggested by Ferrell [
14], we are reporting case studies of any difficult cases.
Demographic measures
Demographic information is collected, including: cognitive functioning (using the Blessed Orientation Memory Concentration test, [
10]), co-morbidity (using the Comorbidity Index and Scores of Charlson, [
15]), performance status (using Karnosfsky scores, [
16]) and ability to perform activities of daily living (Barthel scores, [
17]), age, gender, ethnic group.
Outcome measures for family/friends/recipients of "generativity" documents
In completion of the therapy, semi-structured telephone interviews are conducted with residents' family/friends and other recipients of the "generativity" documents to obtain their perceptions of the impact of the intervention, on themselves and on the resident, and their views on taking part in the study. For residents who die in the data collection period of the study, follow-up telephone interviews are conducted with family/friends/recipients of the "generativity" documents at three months post bereavement. As in the initial interview, these cover their perceptions of the impact of the intervention, on themselves and on the resident, and their views on taking part in the study. They are also asked to complete the Complicated Grief Assessment [
18], which is used to screen for complicated grief reactions. The questionnaire is mailed to them after the telephone interview along with a pre-paid envelope in which to return it. Their age, gender, ethnic group and relationship to residents are also recorded.
Piloting methods for a Phase III trial
Time taken to obtain informed consent and collect outcomes, exclusions, recruitment and drop-out rates (residents and their family and friends) are recorded. In addition their views on taking part in the study (e.g. being randomised) are sought in qualitative interviews, and their views on completing the outcome measures are recorded when they are completing them. Any problems with completing measures will be reported, including missing data.
Analyses
Quantitative data
In view of the relatively small sample size, analyses will be mainly descriptive. However, between and within participant comparisons of outcomes will be conducted and the appropriate effect size estimates reported. The intervention and control group will be compared on the main outcome (a sense of dignity) and secondary outcomes (depression, hopefulness and quality of life). We will also compare Time 2 and Time 3 follow-up with baseline for both groups on these measures. The two groups will also be compared on demographic characteristics and baseline measures. If necessary these will be controlled for in the analyses. We will also report recruitment rates and compare drop-out rates and missing data in the two groups.
Qualitative data
The Framework method of analysis will be used [
19]. Analyses will be both deductive (from pre-set aims and objectives) and inductive (arising from participants views). This method tends to be more structured than some other methods of qualitative analysis and the process more explicit and more informed by a priori questions. It is designed so that it can be more easily understood and assessed by people other than the analyst, such as funding bodies, policy makers and participants. Throughout the analytical process we will use strategies to maximise credibility, criticality, and authenticity[
20]. The QSR NVIVO software package is being used to manage the data.